Students interested in attending the National Conference on Artificial Intelligence in Austin, July 30-August 3, 2000, should consult the AAAI web site for further information about the Student Abstract program and the Doctoral Consortium. Details about these programs have also been mailed to all AAAI members. The Scholarship Program provides partial travel support and a complimentary technical program registration for students who (1) are full-time undergraduate or graduate students at colleges and universities; (2) are members of AAAI; (3) submit papers to the technical program or letters of recommendation from their faculty adviser; and (4) submit scholarship applications to AAAI by April 15, 2000. In addition, repeat scholarship applicants must have fulfilled the volunteer and reporting requirements for previous awards. In the event that scholarship applications AAAI President David Waltz presented The 1999 AAAI Classic Paper Award to exceed available funds, preference John McDermott for R1: An Expert in the Computer Systems Domain.

Human intelligence is shaped by what is most important to us -- the things that cause ecstasy, despair, pleasure, pain, and other intense emotions. The ability to separate the important from the unimportant underlies such faculties as attention, focusing, situation and outcome assessment, priority setting, judgment, taste, goal selection, credit assignment, the selection of relevant memories and precedents, and learning from experience. AI has for the most part focused on logic and reasoning in artificial situations where only relevant variables and operators are specified and has paid insufficient attention to processes of reducing the richness and disorganization of the real world to a form where logical reasoning can be applied. This article discusses the role of importance judgment in intelligence; provides some examples of research that make use of importance judgments; and offers suggestions for new mechanisms, architectures, applications, and research directions for AI.

Because perception-action systems are necessarily constrained by the physics of time and space, robotocists often assume they are best described using differential equations, a language that is specialized for describing the evolution of variables that represent physical quantities. However, when it comes to decision making, where the representations involved refer to goals, strategies, and preferences, AI offers a diverse range of formalisms to the modeler. However, the relationship between these two levels of representation -- signal and symbol -- are not well understood. If we are to achieve success in modeling intelligent physical agents, robotics and AI must reach a new consensus on how to integrate perception-action systems with systems designed for abstract reasoning.

Moravec's estimates of animal equivalence jostling of the atoms in a Moravec's strengths--his insightful are based solely on hardware rock can be seen as the operation of a complete, self-aware mind data analysis, extrapolation of technology complexity. It is often the case that (after Evert) (Everett, H., Many-to extreme conclusions, and hardware alone cannot deliver performance, Worlds of Interpretation/ Quantum provocative predictions--are all here but it also requires software Mechanics, Princeton University and will probably gain him some new sufficient to the task.

It is possible to argue, relatively convincingly, that any research topic only begins to become mature when it appears on a syllabus somewhere. Once the topic has become well enough understood that it can be explained easily to paying customers, and stable enough that anyone teaching it is not likely to have to update his/her teaching materials every few months as new developments are reported, it can be considered to have arrived. Another reasonable indicator of the maturity of a subject, a milestone along the road to academic respectability, is the publication of a really good book on the subject -- not another research monograph but a book that consolidates what is already known, surveys and relates existing ideas, and maybe even unifies some of them. Grigoris Antoniou's Nonmonotonic Reasoning is just such a milestone -- well written, informative, and a good source of information on an important and complex subject. Since the idea was first mooted

The notion of class is ubiquitous in computer science and is central in many formalisms for the representation of structured knowledge used both in knowledge representation and in databases. In this paper we study the basic issues underlying such representation formalisms and single out both their common characteristics and their distinguishing features. Such investigation leads us to propose a unifying framework in which we are able to capture the fundamental aspects of several representation languages used in different contexts. The proposed formalism is expressed in the style of description logics, which have been introduced in knowledge representation as a means to provide a semantically well-founded basis for the structural aspects of knowledge representation systems. The description logic considered in this paper is a subset of first order logic with nice computational characteristics. It is quite expressive and features a novel combination of constructs that has not been studied before. The distinguishing constructs are number restrictions, which generalize existence and functional dependencies, inverse roles, which allow one to refer to the inverse of a relationship, and possibly cyclic assertions, which are necessary for capturing real world domains. We are able to show that it is precisely such combination of constructs that makes our logic powerful enough to model the essential set of features for defining class structures that are common to frame systems, object-oriented database languages, and semantic data models. As a consequence of the established correspondences, several significant extensions of each of the above formalisms become available. The high expressiveness of the logic we propose and the need for capturing the reasoning in different contexts forces us to distinguish between unrestricted and finite model reasoning. A notable feature of our proposal is that reasoning in both cases is decidable. We argue that, by virtue of the high expressive power and of the associated reasoning capabilities on both unrestricted and finite models, our logic provides a common core for class-based representation formalisms.

There are two distinct approaches to solving reinforcement learning problems, namely, searching in value function space and searching in policy space. Temporal difference methods and evolutionary algorithms are well-known examples of these approaches. Kaelbling, Littman and Moore recently provided an informative survey of temporal difference methods. This article focuses on the application of evolutionary algorithms to the reinforcement learning problem, emphasizing alternative policy representations, credit assignment methods, and problem-specific genetic operators. Strengths and weaknesses of the evolutionary approach to reinforcement learning are presented, along with a survey of representative applications.

This paper presents a new approach to identifying and eliminating mislabeled training instances for supervised learning. The goal of this approach is to improve classification accuracies produced by learning algorithms by improving the quality of the training data. Our approach uses a set of learning algorithms to create classifiers that serve as noise filters for the training data. We evaluate single algorithm, majority vote and consensus filters on five datasets that are prone to labeling errors. Our experiments illustrate that filtering significantly improves classification accuracy for noise levels up to 30 percent. An analytical and empirical evaluation of the precision of our approach shows that consensus filters are conservative at throwing away good data at the expense of retaining bad data and that majority filters are better at detecting bad data at the expense of throwing away good data. This suggests that for situations in which there is a paucity of data, consensus filters are preferable, whereas majority vote filters are preferable for situations with an abundance of data.

An ensemble consists of a set of individually trained classifiers (such as neural networks or decision trees) whose predictions are combined when classifying novel instances. Previous research has shown that an ensemble is often more accurate than any of the single classifiers in the ensemble. Bagging (Breiman, 1996c) and Boosting (Freund & Shapire, 1996; Shapire, 1990) are two relatively new but popular methods for producing ensembles. In this paper we evaluate these methods on 23 data sets using both neural networks and decision trees as our classification algorithm. Our results clearly indicate a number of conclusions. First, while Bagging is almost always more accurate than a single classifier, it is sometimes much less accurate than Boosting. On the other hand, Boosting can create ensembles that are less accurate than a single classifier -- especially when using neural networks. Analysis indicates that the performance of the Boosting methods is dependent on the characteristics of the data set being examined. In fact, further results show that Boosting ensembles may overfit noisy data sets, thus decreasing its performance. Finally, consistent with previous studies, our work suggests that most of the gain in an ensemble's performance comes in the first few classifiers combined; however, relatively large gains can be seen up to 25 classifiers when Boosting decision trees.

This article presents a measure of semantic similarity in an IS-A taxonomy based on the notion of shared information content. Experimental evaluation against a benchmark set of human similarity judgments demonstrates that the measure performs better than the traditional edge-counting approach. The article presents algorithms that take advantage of taxonomic similarity in resolving syntactic and semantic ambiguity, along with experimental results demonstrating their effectiveness.